← Search

The Review of Economics and Statistics Vol. 66 No. 2 1984

Comments on Some Properties of X-11

Adi Raveh

Abstract

The Box-Cox technique has been frequently em- ployed in the estimation of economic models for which theory suggests no a priori appropriate functional form. Typically, an iterative OLS procedure is used to find the maximum likelihood estimates of the regression coefficients. However, Spitzer has demonstrated that biased estimates of the coefficient variances result from the use of the OLS covariance matrix. The purpose of this paper is to investigate the magnitude of the bias for two cases typical of the kind encountered in many empirical studies that employ the iterative OLS approach. The paper finds that hypothesis testing based upon the OLS covariance matrix may be quite misleading since the magnitude of the bias is quite large in one of the cases analyzed.

DOI
10.2307/1925840
Volume
66
Issue
2
Pages
343
Sources
openalex crossref

Cite